By splitting the bramble-stumps in the course of July, we perceive also that the Three-pronged Osmia notwithstanding her narrow gallery, follows the same practice as Latreille's Osmia, with a difference. She does not build a party-wall, which the diameter of the cylinder would not permit; she confines herself to putting up a frail circular pad of green putty, as though to limit, before any attempt at harvesting, the space to be occupied by the Bee-bread, whose depth could not be calculated afterwards if the insect did not first mark out its confines. If, in order to see the Osmia's nest as a whole, we split a reed lengthwise, taking care not to disturb its contents; or, better still, if we select for examination the string of cells built in a glass tube, we are forthwith struck by one detail, namely, the uneven distances between the partitions, which are placed almost at right angles to the axis of the cylinder. It is these distances which fix the size of the chambers, which, with a similar base, have different heights and consequently unequal holding-capacities. The bottom partitions, the oldest, are farther apart; those of the front part, near the orifice, are closer together. Moreover, the provisions are plentiful in the loftier cells, whereas they are niggardly and reduced to one-half or even one-third in the cells of lesser height. Let me say at once that the large cells are destined for the females and the small ones for the males. DISTRIBUTION OF THE SEXES. Does the insect which stores up provisions proportionate to the needs of the egg which it is about to lay know beforehand the sex of that egg? Or is the truth even more paradoxical? What we have to do is to turn this suspicion into a certainty demonstrated by experiment. And first let us find out how the sexes are arranged. It is not possible to ascertain the chronological order of a laying, except by going to suitably-chosen species. Fortunately there are a few species in which we do not find this difficulty: these are the Bees who keep to one gallery and build their cells in storeys. Among the number are the different inhabitants of the bramble-stumps, notably the Three-pronged Osmiae, who form an excellent subject for observation, partly because they are of imposing size--bigger than any other bramble-dwellers in my neighbourhood--partly because they are so plentiful.